US20030137349A1 - System and method for a startup circuit for a differential CMOS amplifier - Google Patents
System and method for a startup circuit for a differential CMOS amplifier Download PDFInfo
- Publication number
- US20030137349A1 US20030137349A1 US10/338,918 US33891803A US2003137349A1 US 20030137349 A1 US20030137349 A1 US 20030137349A1 US 33891803 A US33891803 A US 33891803A US 2003137349 A1 US2003137349 A1 US 2003137349A1
- Authority
- US
- United States
- Prior art keywords
- voltage
- circuit
- amplifier
- circuit node
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/45—Differential amplifiers
- H03F3/45071—Differential amplifiers with semiconductor devices only
- H03F3/45479—Differential amplifiers with semiconductor devices only characterised by the way of common mode signal rejection
- H03F3/45632—Differential amplifiers with semiconductor devices only characterised by the way of common mode signal rejection in differential amplifiers with FET transistors as the active amplifying circuit
- H03F3/45695—Differential amplifiers with semiconductor devices only characterised by the way of common mode signal rejection in differential amplifiers with FET transistors as the active amplifying circuit by using feedforward means
- H03F3/4573—Measuring at the common source circuit of the differential amplifier
- H03F3/45739—Controlling the loading circuit of the differential amplifier
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/45—Differential amplifiers
- H03F3/45071—Differential amplifiers with semiconductor devices only
- H03F3/45076—Differential amplifiers with semiconductor devices only characterised by the way of implementation of the active amplifying circuit in the differential amplifier
- H03F3/45179—Differential amplifiers with semiconductor devices only characterised by the way of implementation of the active amplifying circuit in the differential amplifier using MOSFET transistors as the active amplifying circuit
- H03F3/45183—Long tailed pairs
- H03F3/45192—Folded cascode stages
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F3/00—Amplifiers with only discharge tubes or only semiconductor devices as amplifying elements
- H03F3/45—Differential amplifiers
- H03F3/45071—Differential amplifiers with semiconductor devices only
- H03F3/45076—Differential amplifiers with semiconductor devices only characterised by the way of implementation of the active amplifying circuit in the differential amplifier
- H03F3/45376—Differential amplifiers with semiconductor devices only characterised by the way of implementation of the active amplifying circuit in the differential amplifier using junction FET transistors as the active amplifying circuit
- H03F3/45381—Long tailed pairs
- H03F3/4539—Folded cascode stages
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03G—CONTROL OF AMPLIFICATION
- H03G1/00—Details of arrangements for controlling amplification
- H03G1/0005—Circuits characterised by the type of controlling devices operated by a controlling current or voltage signal
- H03G1/0088—Circuits characterised by the type of controlling devices operated by a controlling current or voltage signal using discontinuously variable devices, e.g. switch-operated
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45521—Indexing scheme relating to differential amplifiers the FBC comprising op amp stages, e.g. cascaded stages of the dif amp and being coupled between the LC and the IC
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45594—Indexing scheme relating to differential amplifiers the IC comprising one or more resistors, which are not biasing resistor
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45601—Indexing scheme relating to differential amplifiers the IC comprising one or more passive resistors by feedback
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45616—Indexing scheme relating to differential amplifiers the IC comprising more than one switch, which are not cross coupled
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45698—Indexing scheme relating to differential amplifiers the LC comprising one or more resistors coupled to the LC by feedback (active or passive)
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03F—AMPLIFIERS
- H03F2203/00—Indexing scheme relating to amplifiers with only discharge tubes or only semiconductor devices as amplifying elements covered by H03F3/00
- H03F2203/45—Indexing scheme relating to differential amplifiers
- H03F2203/45726—Indexing scheme relating to differential amplifiers the LC comprising more than one switch, which are not cross coupled
Definitions
- the present invention relates generally to the field of low-noise amplifiers.
- the present invention relates to the development of low-noise programmable gain amplifiers (PGAs) suitable for placement on integrated circuits (ICs) and for use in signal processing applications.
- PGAs low-noise programmable gain amplifiers
- PGAs are used in various analog signal processing applications where an electrical signal of varying amplitude must be either amplified or attenuated before subsequent signal processing.
- Various gain and/or attenuation settings are required to accommodate the wide dynamic range needed for the amplifier's input stages. Numerous conventional techniques exist for meeting these demands.
- the present invention includes a startup circuit including a comparing device configured to (i) receive a second circuit node voltage and a reference voltage as inputs, (ii) compare the received second circuit node voltage and the reference voltage, and (iii) produce a compensating voltage signal based upon the comparison.
- An active device has a control terminal connected to an output port of the comparing device and is configured to receive the compensating voltage signal.
- the active device also includes an output terminal connected to the control terminal of the second active device, and a common terminal connected to a first circuit node.
- Another active device has a control terminal connected to the output port of the comparing device and is configured to receive the compensating voltage signal.
- the other active device also has an output terminal connected to the control terminal of the first active device, and a common terminal connected to the first circuit node.
- FIG. 1 is an illustration of a conventional sub-PGA module having a high input resistance
- FIG. 2 is an illustration of a standard PGA module with active attenuation
- FIG. 3 is a block diagram illustration of an exemplary receive path constructed and arranged in accordance with the present invention.
- FIG. 4 is a more detailed block diagram illustration of a first stage PGA shown in FIG. 3 constructed and arranged in accordance with the present invention
- FIG. 4A is a block diagram illustration used to depict the relationship between impedance, gain, and feedback factor within the first stage PGA of FIG. 4.
- FIG. 5 is a schematic diagram of a conventional common mode feedback circuit which can be used to insure proper start-up conditions in the PGA of FIG. 4;
- FIG. 6 is a block diagram of the amplifier in the PGA of FIG. 4, including an exemplary start-up circuit constructed and arranged in accordance with the present invention
- FIG. 7 is a flow chart of a method of using the start-up circuit of FIG. 6;
- FIG. 8A is an illustration of a first step of a resistor layout constructed and arranged in accordance with the present invention.
- FIG. 8B is an illustration of a second step of the resistor layout of FIG. 8A.
- FIG. 8C is an illustration of a third step of the resistor layouts of FIGS. 8A and 8B.
- FIG. 1 is an illustration of a conventional pre-attenuator 100 followed by a sub-PGA 102 with high-input resistance.
- the pre-attenuator 100 is a separate passive attenuator followed by the sub-PGA 102 , which has a high input resistance.
- a sub-PGA is a circuit block that can provide a variable gain, but does not necessarily provide attenuation.
- the sub-PGA 102 can consist of one or more circuit blocks.
- the sub-PGA 102 may contain a buffer followed by a more traditional PGA module.
- the key characteristic of the sub-PGA 102 is that it typically has high input resistance. Because of the high input resistance of the sub-PGA 102 , the pre-attenuator 100 and sub-PGA 102 do not interact, thus reducing the possibility of a beneficial reduction in the sub-PGA 102 's feedback factor.
- the feedback factor is a numerical index that is related to the ratio of the resistors in closed loop amplifiers that have resistive feedback networks. The ratio of the resistors changes the feedback factor which in-turn changes the amplifier's gain. When amplifier's are used to provide attenuation, the feedback factor typically increases, which consequently makes the amplifier's performance unstable. Therefore, the relationship between gain and the amplifier's feedback factor becomes problematic when the amplifier is used as an attenuator.
- One alternative approach for providing attenuation in amplifiers is to have a passive or an active programmable attenuator followed by a separate, programmable gain amplifier. If the programmable attenuator and the PGA are separate, however, a feedback factor reduction will be difficult to achieve.
- Another alternative approach is to use the programmable gain amplifier as an active attenuator. However, using the programmable gain amplifier as an active attenuator creates two significant problems. First, the issue of the feedback factor becomes a more significant consideration, and the associated substrate area required to accommodate the amplifier and the attenuator become extremely large.
- FIG. 2 a standard PGA circuit 200 with active attenuation is shown in FIG. 2.
- the topology of the PGA circuit 200 there is a standard inverting-gain PGA 202 where attenuation is achieved in the feedback network by making an input resistance 204 larger than a feedback resistance 206 .
- CMOS circuits such as the PGA circuit 200
- the common-mode (CM) signal loop of the first stage amplifier has net positive feedback. Therefore, it is possible that this positive feedback may cause the amplifier to reside in an unwanted latch-up state, especially if the amplifier's positive CM feedback characteristics are able to overpower the negative CM feedback characteristics of the second stage amplifier, also known as the CM feedback amplifier (CMFBA).
- CMFBA CM feedback amplifier
- Another conventional approach for resolving the start-up problem includes providing a high impedance value pull-up resistor to force voltage on a common source (CM-src) node.
- CM-src common source
- a high-valued resistor connected from a supply voltage source V DD to the CM-src node can pull up the CM-src node in order to force proper start-up.
- This approach requires extra power and adds noise to the associated system.
- One aspect of the present invention merges a passive pre-attenuator network with the feedback network used within the amplifier.
- PGA functionality is achieved by using a closed-loop amplifier with a switchable resistor network in the feedback loop to provide passive attenuation. This accomplishes three goals. Since the pre-attenuator can be controlled separately from the rest of the feedback network, it allows for greater controllability of the net PGA gain. Therefore, the total complexity of the feedback network can be reduced.
- the nature of the pre-attenuator of the instant invention is such that when it's used to reduce the overall PGA gain, the overall feedback factor is reduced as well.
- the overall feedback factor increases. Therefore, using the present invention and with careful partitioning of the overall gain between the pre-attenuator and the remainder of the feedback network, one can achieve a wide gain range with a much smaller variation in the feedback factor. This aspect eases the circuit design of the amplifier.
- the first stage feedback network is configured to have CMOS switches placed on a virtual ground. Therefore, no signal current flows through these switches when CMOS technology is used. Also, the switches in the attenautor network have a symmetric differential drive applied to them. Therefore, the linearity of the network will be sufficiently high.
- FIG. 3 provides an illustration of an exemplary PGA 300 , its analog front end, and associated digital circuitry.
- the general purpose of the PGA 300 is to provide gain (amplify) an input signal having a small amplitude so that downstream analog-to-digital converters (ADCs) can receive it and sample a signal having a sufficiently large amplitude. Therefore, the PGA is the first block positioned along a receive path 302 of the PGA 300 .
- the PGA 300 includes three amplification stages 308 , 310 , and 312 .
- the PGA 300 of FIG. 3 can also include an input buffer 314 for driving a switched capacitor circuit (not shown) inside an ADC 315 .
- a low pass filter 316 is included for removing unwanted energy at out-of-band frequencies.
- the input buffer 314 is a fourth amplifier separate and apart from the three amplification stages 308 , 310 , and 312 of the PGA 300 .
- a first aspect of the PGA 300 is associated with switching inside of the PGA, pre-attenuation characteristics, and providing a lower feedback factor.
- FIG. 4 is a more detailed illustration of the first amplification stage 308 of the PGA 300 shown in FIG. 3. Particularly, FIG. 4 shows a resistive feedback network 400 including network segments 401 - 403 (discussed in greater detail below) connected with an amplifier 404 .
- the amplifier 404 is a differential amplifier including differential input and output ports.
- the network segments 401 and 402 are structurally identical. Since the segments 401 and 402 are attached to respective symmetrical differential portions of the amplifier 404 , they are also functionally identical.
- a dotted line 405 indicates an axis of symmetry regarding the layout of the amplifier 404 . Therefore, comments directed to the network segment 401 will also apply to the network segment 402 . Additionally, the network segment 403 is functionally structured along the amplifier's symmetrical axis 405 . The voltage applied to the network segments 401 - 403 is also symmetric and differential. This symmetric differential swing prevents any non-linearities that might be associated with the CMOS process from being excited. Therefore, the linearity of the amplifier 404 and the resistive network 400 will be sufficiently high.
- Gain is a function of the ratio of resistor impedances within the network segments 401 - 403 .
- the network segments 401 - 403 shown in FIG. 4 are connected between non-inverting amplifier input port Vg+ and inverting output port Vo ⁇ and between inverting input port Vg ⁇ and non-inverting output port Vo+ of the amplifier 404 .
- the feedback factor and the gain of the amplifier 404 correspondingly change.
- Changing the feedback factor has a number of effects concerning issues such as the linearity of an amplifier, the noise of the amplifier, and the amplifier's stability.
- the higher the feedback factor the better the performance of the amplifier in terms of noise and linearity.
- the amplifier may lose a measure of stability if the feedback factor is permitted to get too high. In other words, it's more difficult to maintain the stability of the amplifier if its feedback factor is too high.
- the feedback factor is a numerical index derived based upon the topology of the amplifier and the feedback network. It is related to the ratio of the resistor values within the amplifier's associated resistive feedback network. For example, given an amplifier with a resistive feedback network, a user can directly calculate the feedback factor, which will typically be a number between zero and one.
- the feedback factor is also known as the beta-factor of a closed loop amplifier and is more of a quantitative, as opposed to a qualitative, measure of the amplifier's performance. Therefore, changing the ratio of the resistor impedance values changes the feedback factor, which in-turn changes the gain of the amplifier.
- the feedback factor of conventional amplifiers increases and asymptotically approaches the value one, making it harder for the amplifier to become stable.
- the feedback factor reduces and asymptotically approaches the value of zero, and it becomes easier to make the amplifier stable.
- Conventional first stage amplifiers have a wide spread in terms of gain. For example, ranges of about ⁇ 12 dB to +12 dB are not uncommon and are representative of voltage gains of about 0.25 to 4. Thus, the feedback factor can vary by a substantial amount in these conventional amplifiers.
- the resistive network segment 403 is configured to cooperate with the resistive network segments 401 and 402 to facilitate small gains in the amplifier 404 without increasing its feedback factor. More specifically, the network segments 401 and 402 facilitate traditional gain increases within the amplifier 404 .
- resistors 406 a 0 , 406 a 1 - 406 an , and switches 406 b 1 - 406 bn of the network segment 401 cooperatively function to provide the amplifier 404 with gain values greater than or equal to one.
- the resistor 406 a 0 is a first portion of the network segment 401 , while the resistors 406 a 1 - 406 an and switches 406 b 1 - 406 bn collectively represent a second portion of the network segment 401 .
- the exemplary network segment 403 facilitates gain values of less than one.
- the network segment 403 enables the amplifier 404 to attenuate an input signal received at an input port Vin+ of the amplification stage 308 without increasing the amplifier's feedback factor.
- switches 408 b 1 - 408 bm can be closed.
- the switches 408 b 1 - 408 bm provide attenuation for the amplifier 404 and simultaneously lower the feedback factor, consequently improving the amplifier's stability.
- the switches 408 b 1 - 408 bm form a pre-attenuator that enable the amplifier 404 to produce gain values of less than one while also lowering its feedback factor.
- resistors 408 a 1 - 408 am both attenuates the input signal and lowers the amplifier's feedback factor.
- conventional closed loop amplifiers are configured such that the amplifier's attenuation ability and its feedback factor typically operate in opposite directions. In these conventional amplifiers, for example, when an input signal is attenuated, the amplifier's feedback factor typically increases.
- the presence of the switches 408 b 1 - 408 bm and the corresponding resistors 408 a 1 - 408 am allow for gain setting whereby these resistors both attenuate the signal and simultaneously lower the feedback factor.
- This effect is due to the mathematical relationships between the PGA gain and the resistor values in the feedback network.
- the gain and the feedback factor can be directly calculated based upon equations (1) and (2), assuming that the net resistances of each branch of the feedback network are as illustrated in FIG. 4A.
- R 1 , R 2 , and R 4 are controlled based upon which switch of the set 406 b 1 - 406 bn is closed. Also, it is apparent that the value of R 3 is controlled by which switch(es) of the set 408 b 1 - 408 bm is/are closed. In general, as more of the switches of the set 408 b 1 - 408 bm are closed, the value of the R 3 in FIG. 4A will go down.
- the switches 406 b 1 - 406 bn permit the amplifier 404 to achieve a wide variety of gain settings. At any given time, exactly one switch of the set 406 b 1 - 406 bn will be closed. All other switches in this set will be opened. By closing a different switch of this set 406 b 1 - 406 bn , the overall gain of the PGA 308 will be impacted. In general, the closer the switch is to the input nodes (Vin+ and Vin ⁇ ), the higher the resultant gain of the PGA will be when that switch is closed. For example, if switch 406 b 1 were to be closed, that would result in a higher PGA gain that if switch 406 b 4 were to be closed.
- selectively closing and opening the switches 408 b 1 - 408 bm facilitates the achievement of gain values of less than 1 while also lowering the feedback factor.
- the presence of the resistors 408 a 1 - 408 am facilitates both attenuation of the input signal and lowering of the feedback factor.
- the resistors 408 a 1 - 408 am on the opposite sides of the respective switches 408 b 1 - 408 bm are mirror images of each other.
- passive attenuation characteristics are inherently part of the structure of the amplifier 404 and the resistive network segments 401 - 403 that form the amplifier's feedback network 400 .
- the amplifier 404 and the network segments 401 - 403 are completely integrated in terms of their structure and function. That is, the impedances associated, for example, with the switches 408 b 1 - 408 bm are electrically coupled to a feedback path 409 of the amplifier 404 . The result of this coupling is that an attenuation matrix formed within the feedback network 400 cannot be analyzed separately from the gain aspect of the amplifier 404 .
- the network segment 403 is built into the structure of the amplifier 404 and is inherently part of the feedback network 400 .
- FIG. 5 is an illustration of a conventional common mode feedback circuit 500 used to insure proper start-up conditions.
- the conventional circuit 500 is configured to accommodate most common-mode excursions and fluctuations found under normal operating conditions. In essence, the circuit 500 essentially operates as a common-mode feedback circuit correcting typical start-up deficiencies that might impact an amplifier's performance. These typical start up deficiencies, however, are not severe enough to render the amplifier inoperable.
- a differential output signal provided at the output terminals Vo ⁇ and Vo+ of the amplifier 404 in FIG. 4 is received at the input terminals Vo ⁇ and Vo+ of the circuit 500 .
- a resistor string 501 connecting the input terminals Vo ⁇ and Vo+, averages the two voltages as Vcmout so that it becomes the common-mode output of the amplifier.
- Vcmout is then compared with an internally generated reference voltage Vcmref within differential pair transistors 502 and 504 . If Vcmout is higher than Vcmref, then the differential transistor pair 502 / 504 is tilted so that more current flows through transistor 502 than transistor 504 . This pulls a common-mode feedback voltage Vcmfbp at a common-mode feedback transistor 506 , low.
- the circuit 500 pull the common-mode feedback voltage Vcmfbp low.
- Vcmfbp When Vcmfbp is pulled low, then Vo ⁇ and Vo+ are also pulled low via a correcting signal provided at common-mode outputs 506 and 508 , which is in-turn provided as an input to the amplifier 404 .
- the circuit 500 is therefore effective at correcting minor start-up deficiencies in the amplifier 404 such as fluctuations in the common-mode voltage Vcmfbp.
- the circuit 500 has a somewhat limited range and capability. That is, although its effective against, for example, minor common-mode fluctuations, it is not effective at eliminating more severe start-up deficiencies such as common-mode latch-up, which can render the amplifier 404 completely inoperable.
- FIG. 6 provides a more detailed illustration of the amplifier 404 shown in FIG. 4, including an exemplary start-up circuit 600 configured to eliminate the more severe start-up deficiencies such as common-mode latch-up, which can render the amplifier 404 inoperable.
- the input CM, the output CM, and the CM of the gate of the amplifier 404 are significant structural factors. That is, the input to the resistor network, the input to the amplifier, and the output to the amplifier are all dependent upon one another. There are no additional DC paths connected to the input of the PGA 308 (nodes Vin+ and Vin ⁇ of FIG. 4). Therefore, from the DC standpoint, the input to the PGA 308 is floated.
- the amplifier when the input voltage ramps up, could start up in a state where, if the output-common mode is low, for example near ground, then the input to the amplifier will also be pulled down to ground. That is, the input to the amplifier will be substantially the same voltage as the output to the amplifier. Therefore, on start-up, it cannot be certain as to which voltage the amplifier will assume when it is initially powered up, since this cannot be easily controlled. Therefore, if the output-common mode happens to be very low, the input-common mode will also be very low, and that will shut off the amplifier.
- the exemplary start-up circuit 600 senses whether the aforementioned or a similar start-up problem has manifested itself, and then forces the output node of the amplifier 404 to rail. That is, the start-up circuit 600 essentially rails it to V DD , therefore bringing the amplifier out of the bad start-up condition, then turns the start-up circuit 600 off. After operation of the start-up circuit 600 , the amplifieir 404 will then assume a more suitable state of operation. The nature of the amplifier 404 is such that if the output-common mode is too high, that is operating near V DD rail, then the amplifier 404 is still functional, and is devoid of voltage-related start-up problems. On the other hand, if the output-common mode of the amplifier 404 is too low, the amplifier 404 will not start up.
- the differential input ports Vg+ and Vg ⁇ , the differential output ports Vo+ and Vo ⁇ , and voltage sources Vb 1 -Vb 5 of the amplifier 404 are shown.
- Source terminals of input differential active devices 601 and 602 form a common-source node (cmsrc).
- the exemplary start-up circuit 600 and active devices 603 - 604 are also included in the embodiment of FIG. 6 .
- the start up circuit 600 includes a comparing device, such as a comparator 606 and active devices 608 and 610 .
- the active devices 601 - 604 , 608 , and 610 are field effect transistors (FETs), although other active device types can be used.
- FETs field effect transistors
- Common-mode latch-up for example, can pull the differential output terminal Vo+ and Vo ⁇ to ground.
- the comparator 606 monitors a voltage Vcmsrc at the cmsrc node to determine if this voltage goes below a predetermined amount. If Vcmsrc falls below the predetermined amount, the comparator 606 produces an output compensatory voltage Vcmp to a positive supply level to pull up the level of Vo+ and Vo ⁇ . This ultimately pulls Vcmsrc back up, as explained in greater detail below.
- the comparator 606 determines whether Vref is greater than Vcmsrc. If Vref is greater, the comparator 606 outputs the compensatory voltage Vcmp at positive supply. That is, if the positive terminal of the comparator 606 is larger than its negative terminal, a positive supply voltage is produced as an output. This compensatory output voltage Vcmp then turns on the devices 608 and 610 .
- the devices 608 and 610 When the devices 608 and 610 are activated by the compensatory voltage Vcmp, they in-turn pull voltages Vd+ and Vd ⁇ , shown in FIG. 6, basically to ground. Vd+ and Vd ⁇ being pulled to ground temporarily turn off the devices 603 and 604 . This allows Vo+ and Vo ⁇ to be pulled back up. Once Vo+ and Vo ⁇ are pulled back up, Vg+ and Vg ⁇ are pulled up as well through the resistive feedback network 400 , shown in FIG. 4. Once Vg+ and Vg ⁇ are pulled up, the active devices 601 and 602 begin conducting current, enabling the amplifier 404 to reach a stable start-up state. Furthermore, once Vg+ and Vg ⁇ are pulled up, Vcmsrc is also pulled up. When Vcmsrc exceeds Vref, the compensatory output voltage Vcmp gets pulled to ground so that the devices 608 and 610 are turned off. Consequently, the amplifier 404 returns to a normal operation mode.
- FIG. 7 is an illustration of an exemplary method 700 of practicing the present invention.
- a comparing device in a system functionally analogous to the system of FIG. 6 will compare the common-mode source voltage Vcmsrc to the reference voltage Vref in block 702 .
- a compensating voltage Vcmp is produced as an output to the comparing device, as depicted in block 704 .
- the voltages Vo+ and Vo ⁇ are adjusted in accordance with the compensating voltage Vcmp, as depicted in a block 706 , and the amplifier then returns to the normal operating state.
- FIGS. 8 A- 8 C are an illustration of a resistor layout constructed and arranged in accordance with another aspect of the present invention. More specifically, FIGS. 8 A- 8 C depict a technique for geometrically positioning resistors R 1 -R 4 around the PGA to insure matching impedances across the associated IC.
- the layout technique can be used, for example, in construction of the resistive network 400 discussed above and can be implemented using standard IC chip fabrication processes, material, and equipment.
- resistor layout due to IC manufacturing process variations, a mis-match between components, for example resistors or resistances across the IC, may result.
- resistors of equal impedance match substantially well across the entire substrate of the IC.
- each of the resistors R 1 -R 4 is representative of a single resistor value.
- the value of each of the resistors R 1 -R 4 is split to form a number of corresponding respective resistor values R 1 ′-R 4 ′ as shown in FIG. 8B.
- the resistors as shown in FIG. 8C (i.e., forming an interdigital structure across the substrate), substantially equal impedance values can be achieved throughout all the resistors. That is, the geometric pattern established by the resistors R 1 ′-R 4 ′ connected along points A, B and C, and along points D, E and F, of FIG. 8C provide a means of achieving substantially equal impedance values throughout all of the resistors.
- FIGS. 8 A- 8 C The arrangement shown in FIGS. 8 A- 8 C is a type of a common centroid layout, including a technique of splitting resistance values among multiple resistors having intertwining paths. Although two series resistors are illustrated in FIG. 8B, in practice the number of resistors can be extended to any suitable number needed to meet performance requirements. Additionally, the arrangement illustrated in the embodiment of FIG. 8B can also be extended to more parallel paths snaking through the substrate and then recombining in an appropriate fashion as indicated in FIG. 8C.
- the resistors R 1 ′ between points A and B form two parallel paths. That is, a single path begins at point A, diverges, then recombines at point B.
- a feature of the present embodiment is that N number of parallel paths can be split to cover different areas of a substrate so that when they recombine, the device variation over the geometry of the substrate cancels itself out. Therefore, the net result of the technique of FIGS. 8 A- 8 B is an interdigital device with averaged impedances as opposed to a skewed device.
Abstract
Description
- This application is a continuation of the U.S. Non-Provisional Application entitled “System and Method for a Startup Circuit for a Differential CMOS Amplifier,” Ser. No. 10/208,044, filed Jul. 31, 2002, which claims the benefit of U.S. Provisional Application No. 60/350,035, filed Jan. 23, 2002, entitled “System and Method for a Programmable Gain Amplifier,” all of which are incorporated herein in their entireties by reference.
- 1. Field of the Invention
- The present invention relates generally to the field of low-noise amplifiers. In particular, the present invention relates to the development of low-noise programmable gain amplifiers (PGAs) suitable for placement on integrated circuits (ICs) and for use in signal processing applications.
- 2. Related Art
- PGAs are used in various analog signal processing applications where an electrical signal of varying amplitude must be either amplified or attenuated before subsequent signal processing. Various gain and/or attenuation settings are required to accommodate the wide dynamic range needed for the amplifier's input stages. Numerous conventional techniques exist for meeting these demands.
- What are needed, however, are techniques for providing attenuation in closed loop amplifiers without increasing their feedback factor. What is also needed is an approach to ensure suitable start-up conditions and avoid latchup, particularly in complimentary metal oxide semiconductor (CMOS) PGAs. Finally, a technique is needed to eliminate mismatched characteristics commonly found in passive elements across IC substrates due to process gradients.
- The present invention includes a startup circuit including a comparing device configured to (i) receive a second circuit node voltage and a reference voltage as inputs, (ii) compare the received second circuit node voltage and the reference voltage, and (iii) produce a compensating voltage signal based upon the comparison. An active device has a control terminal connected to an output port of the comparing device and is configured to receive the compensating voltage signal. The active device also includes an output terminal connected to the control terminal of the second active device, and a common terminal connected to a first circuit node. Another active device has a control terminal connected to the output port of the comparing device and is configured to receive the compensating voltage signal. The other active device also has an output terminal connected to the control terminal of the first active device, and a common terminal connected to the first circuit node.
- The accompanying drawings, which are incorporated in and constitute part of the specification, illustrate embodiments of the invention and, together with the general description given above and detailed description of the embodiments given below, serve to explain the principles of the present invention.
- FIG. 1 is an illustration of a conventional sub-PGA module having a high input resistance;
- FIG. 2 is an illustration of a standard PGA module with active attenuation;
- FIG. 3 is a block diagram illustration of an exemplary receive path constructed and arranged in accordance with the present invention;
- FIG. 4 is a more detailed block diagram illustration of a first stage PGA shown in FIG. 3 constructed and arranged in accordance with the present invention;
- FIG. 4A is a block diagram illustration used to depict the relationship between impedance, gain, and feedback factor within the first stage PGA of FIG. 4.
- FIG. 5 is a schematic diagram of a conventional common mode feedback circuit which can be used to insure proper start-up conditions in the PGA of FIG. 4;
- FIG. 6 is a block diagram of the amplifier in the PGA of FIG. 4, including an exemplary start-up circuit constructed and arranged in accordance with the present invention;
- FIG. 7 is a flow chart of a method of using the start-up circuit of FIG. 6;
- FIG. 8A is an illustration of a first step of a resistor layout constructed and arranged in accordance with the present invention;
- FIG. 8B is an illustration of a second step of the resistor layout of FIG. 8A; and
- FIG. 8C is an illustration of a third step of the resistor layouts of FIGS. 8A and 8B.
- The following detailed description of the accompanying drawings illustrates exemplary embodiments consistent with the present invention. Other inventions are possible, and modifications may be made to the embodiments within the spirit and scope of the invention. Therefore, the following detailed description is not meant to limit the invention. Rather, the scope of the invention is defined by the appended claims.
- It would be apparent to one of skill in the art that the present invention, as described below, may be implemented in many different embodiments of hardware, software, firmware and/or the entities illustrated in the figures. Any actual software code with the specialized control hardware to implement the present invention, is not limiting of the present invention. Thus, the operation and behavior of the present invention will be described with the understanding that modifications and variations of the embodiments are possible, given the level of detail presented herein.
- FIG. 1 is an illustration of a conventional pre-attenuator100 followed by a
sub-PGA 102 with high-input resistance. In this topology, the pre-attenuator 100 is a separate passive attenuator followed by thesub-PGA 102, which has a high input resistance. A sub-PGA is a circuit block that can provide a variable gain, but does not necessarily provide attenuation. Thesub-PGA 102 can consist of one or more circuit blocks. In particular, thesub-PGA 102 may contain a buffer followed by a more traditional PGA module. - The key characteristic of the
sub-PGA 102 is that it typically has high input resistance. Because of the high input resistance of thesub-PGA 102, the pre-attenuator 100 andsub-PGA 102 do not interact, thus reducing the possibility of a beneficial reduction in thesub-PGA 102's feedback factor. The feedback factor is a numerical index that is related to the ratio of the resistors in closed loop amplifiers that have resistive feedback networks. The ratio of the resistors changes the feedback factor which in-turn changes the amplifier's gain. When amplifier's are used to provide attenuation, the feedback factor typically increases, which consequently makes the amplifier's performance unstable. Therefore, the relationship between gain and the amplifier's feedback factor becomes problematic when the amplifier is used as an attenuator. - One alternative approach for providing attenuation in amplifiers is to have a passive or an active programmable attenuator followed by a separate, programmable gain amplifier. If the programmable attenuator and the PGA are separate, however, a feedback factor reduction will be difficult to achieve. Another alternative approach is to use the programmable gain amplifier as an active attenuator. However, using the programmable gain amplifier as an active attenuator creates two significant problems. First, the issue of the feedback factor becomes a more significant consideration, and the associated substrate area required to accommodate the amplifier and the attenuator become extremely large.
- Next, a
standard PGA circuit 200 with active attenuation is shown in FIG. 2. In the topology of thePGA circuit 200, there is a standard inverting-gain PGA 202 where attenuation is achieved in the feedback network by making aninput resistance 204 larger than afeedback resistance 206. - When using CMOS devices, there is a need to guarantee that the PGA circuits such as the
PGA circuit 200, start up and achieve desirable operating points under a wide variety of start-up, bias, and environmental conditions. In 2-stage differential amplifiers that are typically associated with CMOS circuits, the common-mode (CM) signal loop of the first stage amplifier has net positive feedback. Therefore, it is possible that this positive feedback may cause the amplifier to reside in an unwanted latch-up state, especially if the amplifier's positive CM feedback characteristics are able to overpower the negative CM feedback characteristics of the second stage amplifier, also known as the CM feedback amplifier (CMFBA). - Conventional approaches for eliminating the start-up problem include making the CMFBA large enough so that the primary amp CM loop (positive feedback) can never overpower the CMFBA loop (negative feedback). This approach might be considered a high-power, high-noise solution.
- Another conventional approach for resolving the start-up problem includes providing a high impedance value pull-up resistor to force voltage on a common source (CM-src) node. For an amplifier configured in this manner, a high-valued resistor connected from a supply voltage source VDD to the CM-src node can pull up the CM-src node in order to force proper start-up. This approach, however, requires extra power and adds noise to the associated system.
- Another challenge to the development of PGAs suitable for placement on ICs is that the ICs often require passive elements, such as resistors, that are well-matched to one another. Silicon processing, for example, is an imperfect process that frequently results in process gradients, where the characteristics of a particular device will vary roughly linearly across a certain dimension of the IC's substrate. This linear gradient can cause severe mismatch between devices, such as resistors, that need to be well-matched to one another.
- In order to achieve a sufficient gain range and gain resolution (i.e., gain step size), complex resistive feedback networks are typically used within the amplification modules. This comes at the cost of increased associated die sizes. Additionally, since closed-loop gain and feedback factor are inversely related, in order to achieve a large gain range, it often requires the feedback factor to span a wide range. This complicates the design of the amplifier, which must be guaranteed to be stable and functional over the wide feedback factor span. All these problems must be solved while keeping the closed-loop performance of the PGA sufficiently linear.
- One aspect of the present invention merges a passive pre-attenuator network with the feedback network used within the amplifier. PGA functionality is achieved by using a closed-loop amplifier with a switchable resistor network in the feedback loop to provide passive attenuation. This accomplishes three goals. Since the pre-attenuator can be controlled separately from the rest of the feedback network, it allows for greater controllability of the net PGA gain. Therefore, the total complexity of the feedback network can be reduced.
- Next, the nature of the pre-attenuator of the instant invention is such that when it's used to reduce the overall PGA gain, the overall feedback factor is reduced as well. By contrast, if a conventional feedback network is used to reduce the overall PGA gain, the overall feedback factor increases. Therefore, using the present invention and with careful partitioning of the overall gain between the pre-attenuator and the remainder of the feedback network, one can achieve a wide gain range with a much smaller variation in the feedback factor. This aspect eases the circuit design of the amplifier.
- Finally, the first stage feedback network is configured to have CMOS switches placed on a virtual ground. Therefore, no signal current flows through these switches when CMOS technology is used. Also, the switches in the attenautor network have a symmetric differential drive applied to them. Therefore, the linearity of the network will be sufficiently high.
- FIG. 3 provides an illustration of an
exemplary PGA 300, its analog front end, and associated digital circuitry. The general purpose of thePGA 300 is to provide gain (amplify) an input signal having a small amplitude so that downstream analog-to-digital converters (ADCs) can receive it and sample a signal having a sufficiently large amplitude. Therefore, the PGA is the first block positioned along a receivepath 302 of thePGA 300. Although any number of PGA stages can be accommodated, for purposes of illustration, thePGA 300 includes threeamplification stages - The
PGA 300 of FIG. 3 can also include aninput buffer 314 for driving a switched capacitor circuit (not shown) inside anADC 315. Next, alow pass filter 316 is included for removing unwanted energy at out-of-band frequencies. Theinput buffer 314 is a fourth amplifier separate and apart from the threeamplification stages PGA 300. A first aspect of thePGA 300 is associated with switching inside of the PGA, pre-attenuation characteristics, and providing a lower feedback factor. - FIG. 4 is a more detailed illustration of the
first amplification stage 308 of thePGA 300 shown in FIG. 3. Particularly, FIG. 4 shows aresistive feedback network 400 including network segments 401-403 (discussed in greater detail below) connected with anamplifier 404. In the exemplary embodiment of FIG. 4, theamplifier 404 is a differential amplifier including differential input and output ports. Thenetwork segments 401 and 402 are structurally identical. Since thesegments 401 and 402 are attached to respective symmetrical differential portions of theamplifier 404, they are also functionally identical. - A dotted
line 405 indicates an axis of symmetry regarding the layout of theamplifier 404. Therefore, comments directed to thenetwork segment 401 will also apply to the network segment 402. Additionally, thenetwork segment 403 is functionally structured along the amplifier'ssymmetrical axis 405. The voltage applied to the network segments 401-403 is also symmetric and differential. This symmetric differential swing prevents any non-linearities that might be associated with the CMOS process from being excited. Therefore, the linearity of theamplifier 404 and theresistive network 400 will be sufficiently high. - Another important function of the
PGA 300 is to provide an amplifier having a settable or programmable gain. Progammability can be achieved by altering the input and output impedances of thefeedback network 400. Gain, for example, is a function of the ratio of resistor impedances within the network segments 401-403. The network segments 401-403 shown in FIG. 4 are connected between non-inverting amplifier input port Vg+ and inverting output port Vo− and between inverting input port Vg− and non-inverting output port Vo+ of theamplifier 404. In the process of changing the impedance of the resistors within the network segments 401-403, the feedback factor and the gain of theamplifier 404, correspondingly change. - Changing the feedback factor has a number of effects concerning issues such as the linearity of an amplifier, the noise of the amplifier, and the amplifier's stability. The higher the feedback factor, the better the performance of the amplifier in terms of noise and linearity. However, the amplifier may lose a measure of stability if the feedback factor is permitted to get too high. In other words, it's more difficult to maintain the stability of the amplifier if its feedback factor is too high.
- As noted above, the feedback factor is a numerical index derived based upon the topology of the amplifier and the feedback network. It is related to the ratio of the resistor values within the amplifier's associated resistive feedback network. For example, given an amplifier with a resistive feedback network, a user can directly calculate the feedback factor, which will typically be a number between zero and one. The feedback factor is also known as the beta-factor of a closed loop amplifier and is more of a quantitative, as opposed to a qualitative, measure of the amplifier's performance. Therefore, changing the ratio of the resistor impedance values changes the feedback factor, which in-turn changes the gain of the amplifier.
- In traditional approaches, there is a one-to-one correspondence between the desired gain and the feedback factor present in the amplifier. They are essentially inversely related. The terms are not directly inversely related because the associated mathematical expressions depict a more complicated relationship.
- If one desires a low gain or equivalently, desires attenuation, the feedback factor of conventional amplifiers increases and asymptotically approaches the value one, making it harder for the amplifier to become stable. On the other hand, if one desires a higher gain, the feedback factor reduces and asymptotically approaches the value of zero, and it becomes easier to make the amplifier stable. Conventional first stage amplifiers have a wide spread in terms of gain. For example, ranges of about −12 dB to +12 dB are not uncommon and are representative of voltage gains of about 0.25 to 4. Thus, the feedback factor can vary by a substantial amount in these conventional amplifiers.
- In the present invention, however, the
resistive network segment 403 is configured to cooperate with theresistive network segments 401 and 402 to facilitate small gains in theamplifier 404 without increasing its feedback factor. More specifically, thenetwork segments 401 and 402 facilitate traditional gain increases within theamplifier 404. For example, resistors 406 a 0, 406 a 1-406 an, and switches 406 b 1-406 bn of thenetwork segment 401, cooperatively function to provide theamplifier 404 with gain values greater than or equal to one. The resistor 406 a 0 is a first portion of thenetwork segment 401, while the resistors 406 a 1-406 an and switches 406 b 1-406 bn collectively represent a second portion of thenetwork segment 401. - The
exemplary network segment 403, on the other hand, facilitates gain values of less than one. In particular, thenetwork segment 403 enables theamplifier 404 to attenuate an input signal received at an input port Vin+ of theamplification stage 308 without increasing the amplifier's feedback factor. When attenuation is desired, switches 408 b 1-408 bm can be closed. The switches 408 b 1-408 bm provide attenuation for theamplifier 404 and simultaneously lower the feedback factor, consequently improving the amplifier's stability. In other words, the switches 408 b 1-408 bm form a pre-attenuator that enable theamplifier 404 to produce gain values of less than one while also lowering its feedback factor. - Conversely, the approach of the present exemplary embodiment of FIG. 4 makes it easier to design amplifiers without concern for stability related issues. The presence of resistors408 a 1-408 am both attenuates the input signal and lowers the amplifier's feedback factor. As noted above, conventional closed loop amplifiers are configured such that the amplifier's attenuation ability and its feedback factor typically operate in opposite directions. In these conventional amplifiers, for example, when an input signal is attenuated, the amplifier's feedback factor typically increases.
-
- In FIG. 4A, it is apparent that the values of R1, R2, and R4 are controlled based upon which switch of the set 406 b 1-406 bn is closed. Also, it is apparent that the value of R3 is controlled by which switch(es) of the set 408 b 1-408 bm is/are closed. In general, as more of the switches of the set 408 b 1-408 bm are closed, the value of the R3 in FIG. 4A will go down.
- Examining the equations (1) and (2), one can directly calculate that reducing the value of R3 simultaneously lowers the gain and the feedback factor. Therefore, it can equivalently be stated that: by closing one or more of the set 408 b 1-408 bm, both the gain and the feedback factor are reduced.
- The switches406 b 1-406 bn permit the
amplifier 404 to achieve a wide variety of gain settings. At any given time, exactly one switch of the set 406 b 1-406 bn will be closed. All other switches in this set will be opened. By closing a different switch of this set 406 b 1-406 bn, the overall gain of thePGA 308 will be impacted. In general, the closer the switch is to the input nodes (Vin+ and Vin−), the higher the resultant gain of the PGA will be when that switch is closed. For example, if switch 406b 1 were to be closed, that would result in a higher PGA gain that if switch 406 b 4 were to be closed. - Similarly, selectively closing and opening the switches408 b 1-408 bm facilitates the achievement of gain values of less than 1 while also lowering the feedback factor. The presence of the resistors 408 a 1-408 am facilitates both attenuation of the input signal and lowering of the feedback factor. The resistors 408 a 1-408 am on the opposite sides of the respective switches 408 b 1-408 bm are mirror images of each other.
- Also, in the present invention, passive attenuation characteristics are inherently part of the structure of the
amplifier 404 and the resistive network segments 401-403 that form the amplifier'sfeedback network 400. Thus, theamplifier 404 and the network segments 401-403 are completely integrated in terms of their structure and function. That is, the impedances associated, for example, with the switches 408 b 1-408 bm are electrically coupled to afeedback path 409 of theamplifier 404. The result of this coupling is that an attenuation matrix formed within thefeedback network 400 cannot be analyzed separately from the gain aspect of theamplifier 404. More precisely, thenetwork segment 403 is built into the structure of theamplifier 404 and is inherently part of thefeedback network 400. - FIG. 5 is an illustration of a conventional common
mode feedback circuit 500 used to insure proper start-up conditions. As previously noted, particularly in CMOS circuits, there is a need to guarantee that the PGA circuit starts up and achieves a desirable operating point under number of environmental conditions. Theconventional circuit 500 is configured to accommodate most common-mode excursions and fluctuations found under normal operating conditions. In essence, thecircuit 500 essentially operates as a common-mode feedback circuit correcting typical start-up deficiencies that might impact an amplifier's performance. These typical start up deficiencies, however, are not severe enough to render the amplifier inoperable. - In the
circuit 500, a differential output signal provided at the output terminals Vo− and Vo+ of theamplifier 404 in FIG. 4 is received at the input terminals Vo− and Vo+ of thecircuit 500. Aresistor string 501, connecting the input terminals Vo− and Vo+, averages the two voltages as Vcmout so that it becomes the common-mode output of the amplifier. Vcmout is then compared with an internally generated reference voltage Vcmref withindifferential pair transistors differential transistor pair 502/504 is tilted so that more current flows throughtransistor 502 thantransistor 504. This pulls a common-mode feedback voltage Vcmfbp at a common-mode feedback transistor 506, low. - Thus, if the common-mode output is too high, the
circuit 500 pull the common-mode feedback voltage Vcmfbp low. When Vcmfbp is pulled low, then Vo− and Vo+ are also pulled low via a correcting signal provided at common-mode outputs amplifier 404. Thecircuit 500 is therefore effective at correcting minor start-up deficiencies in theamplifier 404 such as fluctuations in the common-mode voltage Vcmfbp. Thecircuit 500, however, has a somewhat limited range and capability. That is, although its effective against, for example, minor common-mode fluctuations, it is not effective at eliminating more severe start-up deficiencies such as common-mode latch-up, which can render theamplifier 404 completely inoperable. - FIG. 6 provides a more detailed illustration of the
amplifier 404 shown in FIG. 4, including an exemplary start-upcircuit 600 configured to eliminate the more severe start-up deficiencies such as common-mode latch-up, which can render theamplifier 404 inoperable. - In the illustrations of FIGS. 4 and 6, the input CM, the output CM, and the CM of the gate of the
amplifier 404 are significant structural factors. That is, the input to the resistor network, the input to the amplifier, and the output to the amplifier are all dependent upon one another. There are no additional DC paths connected to the input of the PGA 308 (nodes Vin+ and Vin− of FIG. 4). Therefore, from the DC standpoint, the input to thePGA 308 is floated. - From a start-up perspective, the amplifier, when the input voltage ramps up, could start up in a state where, if the output-common mode is low, for example near ground, then the input to the amplifier will also be pulled down to ground. That is, the input to the amplifier will be substantially the same voltage as the output to the amplifier. Therefore, on start-up, it cannot be certain as to which voltage the amplifier will assume when it is initially powered up, since this cannot be easily controlled. Therefore, if the output-common mode happens to be very low, the input-common mode will also be very low, and that will shut off the amplifier.
- In the illustrative embodiment of FIG. 6, the exemplary start-up
circuit 600 senses whether the aforementioned or a similar start-up problem has manifested itself, and then forces the output node of theamplifier 404 to rail. That is, the start-upcircuit 600 essentially rails it to VDD, therefore bringing the amplifier out of the bad start-up condition, then turns the start-upcircuit 600 off. After operation of the start-upcircuit 600, theamplifieir 404 will then assume a more suitable state of operation. The nature of theamplifier 404 is such that if the output-common mode is too high, that is operating near VDD rail, then theamplifier 404 is still functional, and is devoid of voltage-related start-up problems. On the other hand, if the output-common mode of theamplifier 404 is too low, theamplifier 404 will not start up. - In the illustration of FIG. 6, the differential input ports Vg+ and Vg−, the differential output ports Vo+ and Vo−, and voltage sources Vb1-Vb5 of the
amplifier 404 are shown. Source terminals of input differentialactive devices circuit 600 and active devices 603-604. - The start up
circuit 600 includes a comparing device, such as acomparator 606 andactive devices comparator 606 monitors a voltage Vcmsrc at the cmsrc node to determine if this voltage goes below a predetermined amount. If Vcmsrc falls below the predetermined amount, thecomparator 606 produces an output compensatory voltage Vcmp to a positive supply level to pull up the level of Vo+ and Vo−. This ultimately pulls Vcmsrc back up, as explained in greater detail below. - In monitoring Vcmsrc, the
comparator 606 determines whether Vref is greater than Vcmsrc. If Vref is greater, thecomparator 606 outputs the compensatory voltage Vcmp at positive supply. That is, if the positive terminal of thecomparator 606 is larger than its negative terminal, a positive supply voltage is produced as an output. This compensatory output voltage Vcmp then turns on thedevices - When the
devices devices resistive feedback network 400, shown in FIG. 4. Once Vg+ and Vg− are pulled up, theactive devices amplifier 404 to reach a stable start-up state. Furthermore, once Vg+ and Vg− are pulled up, Vcmsrc is also pulled up. When Vcmsrc exceeds Vref, the compensatory output voltage Vcmp gets pulled to ground so that thedevices amplifier 404 returns to a normal operation mode. - FIG. 7 is an illustration of an
exemplary method 700 of practicing the present invention. In FIG. 7, a comparing device in a system functionally analogous to the system of FIG. 6 will compare the common-mode source voltage Vcmsrc to the reference voltage Vref inblock 702. As a result of the comparison, a compensating voltage Vcmp is produced as an output to the comparing device, as depicted inblock 704. Finally, the voltages Vo+ and Vo− are adjusted in accordance with the compensating voltage Vcmp, as depicted in ablock 706, and the amplifier then returns to the normal operating state. - FIGS.8A-8C are an illustration of a resistor layout constructed and arranged in accordance with another aspect of the present invention. More specifically, FIGS. 8A-8C depict a technique for geometrically positioning resistors R1-R4 around the PGA to insure matching impedances across the associated IC. The layout technique can be used, for example, in construction of the
resistive network 400 discussed above and can be implemented using standard IC chip fabrication processes, material, and equipment. - The problem addressed by the resistor layout is that due to IC manufacturing process variations, a mis-match between components, for example resistors or resistances across the IC, may result. In IC manufacturing, it is desirable that resistors of equal impedance match substantially well across the entire substrate of the IC.
- In the FIG. 8A, each of the resistors R1-R4 is representative of a single resistor value. First, the value of each of the resistors R1-R4 is split to form a number of corresponding respective resistor values R1′-R4′ as shown in FIG. 8B. By then configuring the resistors as shown in FIG. 8C (i.e., forming an interdigital structure across the substrate), substantially equal impedance values can be achieved throughout all the resistors. That is, the geometric pattern established by the resistors R1′-R4′ connected along points A, B and C, and along points D, E and F, of FIG. 8C provide a means of achieving substantially equal impedance values throughout all of the resistors.
- The arrangement shown in FIGS.8A-8C is a type of a common centroid layout, including a technique of splitting resistance values among multiple resistors having intertwining paths. Although two series resistors are illustrated in FIG. 8B, in practice the number of resistors can be extended to any suitable number needed to meet performance requirements. Additionally, the arrangement illustrated in the embodiment of FIG. 8B can also be extended to more parallel paths snaking through the substrate and then recombining in an appropriate fashion as indicated in FIG. 8C.
- In FIG. 8C, as noted, the resistors R1′ between points A and B form two parallel paths. That is, a single path begins at point A, diverges, then recombines at point B. A feature of the present embodiment is that N number of parallel paths can be split to cover different areas of a substrate so that when they recombine, the device variation over the geometry of the substrate cancels itself out. Therefore, the net result of the technique of FIGS. 8A-8B is an interdigital device with averaged impedances as opposed to a skewed device.
- The foregoing description of the preferred embodiments provide an illustration and description, but is not intended to be exhaustive or to limit the invention to the precise form disclosed. Modifications and variations are possible consistent with the above teachings or may be acquired from practice of the invention. Thus, it is noted that the scope of the invention is defined by the claims and their equivalents.
Claims (15)
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/338,918 US6653901B2 (en) | 2002-01-23 | 2003-01-09 | System and method for a startup circuit for a differential CMOS amplifier |
US10/664,076 US6891434B2 (en) | 2002-01-23 | 2003-09-17 | System and method for a startup circuit for a differential CMOS amplifier |
US11/103,424 US7061318B2 (en) | 2002-01-23 | 2005-04-12 | System and method for a startup circuit for a differential CMOS amplifier |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US35003502P | 2002-01-23 | 2002-01-23 | |
US10/208,044 US6570448B1 (en) | 2002-01-23 | 2002-07-31 | System and method for a startup circuit for a differential CMOS amplifier |
US10/338,918 US6653901B2 (en) | 2002-01-23 | 2003-01-09 | System and method for a startup circuit for a differential CMOS amplifier |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/208,044 Continuation US6570448B1 (en) | 2002-01-23 | 2002-07-31 | System and method for a startup circuit for a differential CMOS amplifier |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/664,076 Continuation US6891434B2 (en) | 2002-01-23 | 2003-09-17 | System and method for a startup circuit for a differential CMOS amplifier |
Publications (2)
Publication Number | Publication Date |
---|---|
US20030137349A1 true US20030137349A1 (en) | 2003-07-24 |
US6653901B2 US6653901B2 (en) | 2003-11-25 |
Family
ID=26902848
Family Applications (4)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/208,044 Expired - Fee Related US6570448B1 (en) | 2002-01-23 | 2002-07-31 | System and method for a startup circuit for a differential CMOS amplifier |
US10/338,918 Expired - Lifetime US6653901B2 (en) | 2002-01-23 | 2003-01-09 | System and method for a startup circuit for a differential CMOS amplifier |
US10/664,076 Expired - Fee Related US6891434B2 (en) | 2002-01-23 | 2003-09-17 | System and method for a startup circuit for a differential CMOS amplifier |
US11/103,424 Expired - Lifetime US7061318B2 (en) | 2002-01-23 | 2005-04-12 | System and method for a startup circuit for a differential CMOS amplifier |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/208,044 Expired - Fee Related US6570448B1 (en) | 2002-01-23 | 2002-07-31 | System and method for a startup circuit for a differential CMOS amplifier |
Family Applications After (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/664,076 Expired - Fee Related US6891434B2 (en) | 2002-01-23 | 2003-09-17 | System and method for a startup circuit for a differential CMOS amplifier |
US11/103,424 Expired - Lifetime US7061318B2 (en) | 2002-01-23 | 2005-04-12 | System and method for a startup circuit for a differential CMOS amplifier |
Country Status (3)
Country | Link |
---|---|
US (4) | US6570448B1 (en) |
EP (1) | EP1347571B1 (en) |
DE (1) | DE60319297T2 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030128856A1 (en) * | 2002-01-08 | 2003-07-10 | Boor Steven E. | Digitally programmable gain amplifier |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE10125366A1 (en) * | 2001-05-23 | 2002-12-12 | Infineon Technologies Ag | dB linear VGA stage with high bandwidth |
JP3792207B2 (en) * | 2003-03-25 | 2006-07-05 | 沖電気工業株式会社 | Current-driven differential driver and data transmission method using current-driven differential driver |
US7180369B1 (en) | 2003-05-15 | 2007-02-20 | Marvell International Ltd. | Baseband filter start-up circuit |
GB2415844B (en) * | 2004-06-30 | 2006-05-03 | Renesas Tech Corp | Wireless communication receiver |
DE102005036100B4 (en) * | 2005-05-18 | 2012-04-12 | Infineon Technologies Ag | Amplifier arrangement and method for amplifying a voltage |
US7679443B2 (en) * | 2006-08-31 | 2010-03-16 | Texas Instruments Incorporated | System and method for common mode translation |
US8710899B2 (en) * | 2008-09-17 | 2014-04-29 | Lockheed Martin Corporation | Stepped delay control of integrated switches |
EP2498400A1 (en) * | 2011-03-11 | 2012-09-12 | Dialog Semiconductor GmbH | A delta-sigma modulator approach to increased amplifier gain resolution |
US9083297B2 (en) | 2012-05-10 | 2015-07-14 | That Corporation | Programmable-gain amplifier |
US9425149B1 (en) | 2013-11-22 | 2016-08-23 | Altera Corporation | Integrated circuit package routing with reduced crosstalk |
CN107993628B (en) * | 2018-01-26 | 2020-05-12 | 京东方科技集团股份有限公司 | Common voltage compensation circuit, compensation method thereof, display panel and display device |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3859477A (en) * | 1971-06-24 | 1975-01-07 | Tesla Np | Electrostatic transducer |
US4117275A (en) * | 1976-06-11 | 1978-09-26 | Chemi-Con Onkyo Co., Ltd. | Non-directional electret microphone with an air passage to balance pressures on opposite sides of the diaphragm |
US6366678B1 (en) * | 1999-01-07 | 2002-04-02 | Sarnoff Corporation | Microphone assembly for hearing aid with JFET flip-chip buffer |
Family Cites Families (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4024462A (en) * | 1975-05-27 | 1977-05-17 | International Business Machines Corporation | Darlington configuration high frequency differential amplifier with zero offset current |
JPS5573114A (en) * | 1978-11-28 | 1980-06-02 | Nippon Gakki Seizo Kk | Output offset control circuit for full step direct-coupled amplifier |
DE69118693T2 (en) * | 1991-09-25 | 1996-11-28 | Bell Telephone Mfg | Differential amplifier arrangement |
JPH07106875A (en) * | 1993-09-30 | 1995-04-21 | Nec Corp | Semiconductor integrated circuit |
US6107882A (en) * | 1997-12-11 | 2000-08-22 | Lucent Technologies Inc. | Amplifier having improved common mode voltage range |
IT1309712B1 (en) * | 1999-02-19 | 2002-01-30 | St Microelectronics Srl | HIGH-FREQUENCY VARIABLE GAIN MULTI-STAGE AMPLIFIER AND REDUCED PHASE VARIATIONS |
GB2358532A (en) * | 2000-01-22 | 2001-07-25 | Mitel Semiconductor Ltd | AC voltage amplifier using current mode stages |
-
2002
- 2002-07-31 US US10/208,044 patent/US6570448B1/en not_active Expired - Fee Related
-
2003
- 2003-01-09 US US10/338,918 patent/US6653901B2/en not_active Expired - Lifetime
- 2003-01-23 EP EP03001572A patent/EP1347571B1/en not_active Expired - Fee Related
- 2003-01-23 DE DE60319297T patent/DE60319297T2/en not_active Expired - Lifetime
- 2003-09-17 US US10/664,076 patent/US6891434B2/en not_active Expired - Fee Related
-
2005
- 2005-04-12 US US11/103,424 patent/US7061318B2/en not_active Expired - Lifetime
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3859477A (en) * | 1971-06-24 | 1975-01-07 | Tesla Np | Electrostatic transducer |
US4117275A (en) * | 1976-06-11 | 1978-09-26 | Chemi-Con Onkyo Co., Ltd. | Non-directional electret microphone with an air passage to balance pressures on opposite sides of the diaphragm |
US6366678B1 (en) * | 1999-01-07 | 2002-04-02 | Sarnoff Corporation | Microphone assembly for hearing aid with JFET flip-chip buffer |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030128856A1 (en) * | 2002-01-08 | 2003-07-10 | Boor Steven E. | Digitally programmable gain amplifier |
Also Published As
Publication number | Publication date |
---|---|
US6570448B1 (en) | 2003-05-27 |
US20050179491A1 (en) | 2005-08-18 |
US6891434B2 (en) | 2005-05-10 |
US20040051586A1 (en) | 2004-03-18 |
US7061318B2 (en) | 2006-06-13 |
EP1347571A2 (en) | 2003-09-24 |
EP1347571A3 (en) | 2005-02-09 |
DE60319297D1 (en) | 2008-04-10 |
DE60319297T2 (en) | 2009-02-26 |
EP1347571B1 (en) | 2008-02-27 |
US6653901B2 (en) | 2003-11-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US6833759B2 (en) | System and method for a programmable gain amplifier | |
USRE43776E1 (en) | Layout technique for matched resistors on an integrated circuit substrate | |
US7061318B2 (en) | System and method for a startup circuit for a differential CMOS amplifier | |
US6873210B2 (en) | Single-ended-to-differential converter with common-mode voltage control | |
US7821333B2 (en) | High-voltage differential amplifier and method using low voltage amplifier and dynamic voltage selection | |
US7154334B2 (en) | Common mode feedback circuit for fully differential two-stage operational amplifiers | |
US7151409B2 (en) | Programmable low noise amplifier and method | |
US7295070B2 (en) | Flip around switched capacitor amplifier | |
US6642788B1 (en) | Differential cascode amplifier | |
Carrillo et al. | 1-V rail-to-rail bulk-driven CMOS OTA with enhanced gain and gain-bandwidth product | |
US4947135A (en) | Single-ended chopper stabilized operational amplifier | |
US20100295594A1 (en) | Source Follower Attenuator | |
US7057454B2 (en) | Resistor and switch-minimized variable analog gain circuit | |
US10236851B2 (en) | Wide bandwidth variable gain amplifier and exponential function generator | |
JPH04323907A (en) | Low-offset, high-speed cmos amplifier which separates noise in supplied electric power | |
Schlögl et al. | Low-voltage operational amplifier in 0.12 μm digital CMOS technology |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
FPAY | Fee payment |
Year of fee payment: 12 |
|
AS | Assignment |
Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH CAROLINA Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001 Effective date: 20160201 Owner name: BANK OF AMERICA, N.A., AS COLLATERAL AGENT, NORTH Free format text: PATENT SECURITY AGREEMENT;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:037806/0001 Effective date: 20160201 |
|
AS | Assignment |
Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD., SINGAPORE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 Owner name: AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:BROADCOM CORPORATION;REEL/FRAME:041706/0001 Effective date: 20170120 |
|
AS | Assignment |
Owner name: BROADCOM CORPORATION, CALIFORNIA Free format text: TERMINATION AND RELEASE OF SECURITY INTEREST IN PATENTS;ASSIGNOR:BANK OF AMERICA, N.A., AS COLLATERAL AGENT;REEL/FRAME:041712/0001 Effective date: 20170119 |
|
AS | Assignment |
Owner name: AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE. LIMITE Free format text: MERGER;ASSIGNOR:AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD.;REEL/FRAME:047195/0026 Effective date: 20180509 |
|
AS | Assignment |
Owner name: AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE. LIMITE Free format text: CORRECTIVE ASSIGNMENT TO CORRECT THE EFFECTIVE DATE OF MERGER PREVIOUSLY RECORDED ON REEL 047195 FRAME 0026. ASSIGNOR(S) HEREBY CONFIRMS THE MERGER;ASSIGNOR:AVAGO TECHNOLOGIES GENERAL IP (SINGAPORE) PTE. LTD.;REEL/FRAME:047477/0423 Effective date: 20180905 |